Gas Discharge tube-when power up spikes comes out.

Hi All,

I am doing a design using a Gas discharge tube. The Gas discharge tube is the next one:

https://www.mouser.co.uk/ProductDet...GAEpiMZZMsdWQmBKEIh8inVW%2ByC3JoT/R5pQ9LyuD4=

The product is rated to 230V ( it is its normal operating voltage). The problem is that the GDT is showing some sparks at 230V when It should is rated 430V, so there shouldn't be any reason why the GDT is sparking.

Is there any way to absorb those spark, avoid them or to choose another GDT which doesn't spark ( or at least the spark stay contained in the encapsulated).

Cheers,
Winsu

A GDT does not trigger unless the voltage across it exceeds flash-over level but they are very fast to react. I suspect there really is a high voltage pulse in your circuit, possibly from a resonance in wiring or something inductive switching off at the same time.

You should identify the cause but a 'work around' might be to add a small inductance in series with the voltage feed to slow down any fast edged spikes on the wire.

Brian.

If the tube is for protection, given that its avalanche mode, hence very fast,
maybe track down the unwanted source of HV before considering slowing
down the tube and compromising its performance ? Just a thought.

Regards, Dana.

- - - Updated - - -

Have you looked at tube with a DSO set to trigger on worst case trig V of tube.

You are getting transients that exceed the avalanche rating of tube. Eg. its doing
its job.

Thanks for your answers!.

The gas discharge tube is at the very front of a line filter and it is next to a common mode choke so any spike voltage theoretically should be mitigated by the common mode choke ( although it would be good to increase its value and see what happens). The switching stage is very far away from the GDT so I think it is unlike that induction is the source of this problem, but it will be worth to check. I am hoping it is a problem with the power supply otherwise it looks like it will be very difficult to spot.

Another question is related to its expected functionality. The GDT will spark when it triggers, how can I mitigate or get the rid of those spikes( at least the spikes that occur externally the encapsulated). This product should be comply with ATEX and with spikes around I don't think it would pass.

Regards,
Winsu.

Are you in a high humidity environment ? Given that it is arching over exterior of
package. Maybe a "longer" package with so that external arc V >> Gas Avalanche V ?


Regards, Dana.

It is intended to be installed in high humidity environments....

I don't follow you with "arc V >> Gas Avalanche V", I though the arc is how it discharges the energy so the arc voltage should be equal to gas avalanche V.

I got the impression you observed arcing outside the package, between terminal
connections.

The "arc V >> Gas Avalanche V" means that the end cap connections to tube do
not arc, because they are close to each other, before the internal gas avalanche occurs.
Humidity would play a role here as well.


Regards, Dana.

Can you give as a circuit diagram? The 230V rating mentioned in post #1 makes me wonder if you are connecting the GDT across 230V mains voltage. This is likely to damage the device.

The GDT has three terminal, it is at the very front of the input voltage. The central pin is connected to earth, the two remaining pins to the Live and Neutral. It is rated 430V, so it shouldn't be clamping ( as far as I am aware).Is it the correct way to connect it?

Winsu,
Cheers.

These GDT are not suited for mains circuit protection. After being triggered, the GDT shorts the electrode voltage to 10 V. Depending on the mains impedance, the short circuit current will blow the fuse, the GDT or both.

Do you mean that GDTs in general are not suited for mains? or these type specifically are not suited for main applications. If there are some GDT that are suitable could you please point them out?.

Another general doubt that I have about GDT is the next: I know that they triggers and they dissipate the energy through a spark, will that spark always be contained inside the encapsulation as long as the GDT is performing correctly?.